Visit the website of almost any congressional campaign and you’ll encounter a digital experience that would embarrass a 2010 startup. Slow load times. Confusing navigation. Mobile experiences that feel like afterthoughts. Donation flows with unnecessary friction. Email capture forms that look like they were designed by someone who last built a website in 2008.
This is not because the people running these campaigns are technologically unsophisticated. Many of them are. Some campaigns spend significant money on their digital presence.
It’s because the incentive structure of political campaigns does not reward good user experience. It rewards extraction.
What a Campaign Website Is Actually For
Start with what a campaign website is actually optimized to do, and the design choices make immediate sense.
A campaign website has two primary functions: collect email addresses and collect donations. Everything else — candidate biography, policy positions, endorsements, volunteer signup — is secondary infrastructure that serves these two goals.
This is not a cynical observation. It’s an accurate description of how campaign managers and digital directors are evaluated. The success metric is not “did users find the information they were looking for?” It is “did users give us their contact information or their money?”
The success metric is not “did users find the information they were looking for?” It is “did users give us their contact information or their money?”
These two goals are not incompatible with good user experience, but they create a specific optimization pressure that diverges from the optimization pressure in commercial products. A commercial product needs users to come back. Retention matters. Word of mouth matters. The experience needs to be good enough that the user tells someone else about it.
A campaign website has a hard expiration date. There is no return user problem to solve. The user pool is largely fixed — registered voters in a geography — and the mission is to extract maximum value from each visit before Election Day. The incentive to invest in a pleasurable experience, one that the user will recommend to others, is essentially zero.
The Vendor Ecosystem Problem
The structural reason campaign websites are bad is that the vendor ecosystem that builds them was not created to compete on user experience.
Political technology vendors are evaluated by campaigns on a narrow set of criteria: Does it integrate with ActBlue/WinRed? Does it connect to the voter file? Does it have email list management? Does the support team know how to run a campaign? Political experience is treated as a differentiating asset that justifies substantial price premiums and forgives substantial design shortcomings.
The result is a market that has very limited competitive pressure on the dimensions where commercial technology markets apply maximum pressure — speed, design quality, mobile optimization, conversion rate optimization in the user’s interest rather than the campaign’s.
Political technology firms with genuinely bad products survive for years because their clients don’t know what good looks like and don’t have time to find out.
The consultants who recommend technology vendors to campaigns are generally former campaign staffers who know campaigns, not technology professionals who know technology. They evaluate products on political fit, not product quality. And they often have financial relationships with the vendors they recommend.
The Deeper Problem: Government Experience
The campaign website problem is actually a mild version of a much larger dysfunction: the catastrophically bad technology infrastructure that most Americans interact with when they engage with government.
DMV websites. Court filing systems. Benefits portals. Tax filing interfaces. Public records request systems. The technology through which hundreds of millions of Americans interact with government at the federal, state, and local level is, with rare exceptions, significantly worse than the technology they use to order pizza.
This is not a resource problem, at least not primarily. Federal, state, and local governments spend substantial amounts on technology. The problem is procurement — the process by which government buys technology is structured in ways that systematically favor large, incumbent vendors with compliance expertise over small, innovative vendors with good products.
Government technology procurement rewards the vendor who can navigate a complex RFP process, maintain relationships with procurement officials, and produce the documentation required for compliance review. It does not reward the vendor who builds the best product. The incentive structure is essentially the opposite of the commercial market’s.
The result is a technology experience that communicates something to citizens about their relationship with their government whether anyone intended that message or not: your time and experience do not matter here.
The Civic Cost
There is a theory in the UX literature called “friction as a signal.” The idea is that when a product is difficult to use, users internalize a message about how much the product’s creators value their time and engagement.
When government websites are hard to use, when benefits portals time out, when court filing systems require fax machines in 2024, when voting registration interfaces work on desktop but not mobile — the message citizens receive, through the interface rather than through any explicit statement, is that their participation is not genuinely wanted. The friction is a signal.
When government websites are hard to use... the message citizens receive, through the interface rather than through any explicit statement, is that their participation is not genuinely wanted. The friction is a signal.
The research on civic participation and technology quality is consistent: reducing friction in civic participation interfaces increases participation, particularly among young voters and voters who are less civically engaged as a baseline. The inverse is also true — high-friction civic technology suppresses participation among exactly the populations that are most likely to already have lower participation rates.
This is not a technology story. It’s a democracy story told through technology.
The Exception That Proves the Rule
The cases where government and political technology have been done well are instructive precisely because they reveal what’s possible when the incentive structure changes.
The UK’s Government Digital Service, established in 2011, applied commercial UX standards to government interfaces with dramatic results: GOV.UK went from dozens of separate agency websites with divergent design and poor usability to a single, coherent interface with clear information architecture and well-designed user flows. User satisfaction scores increased substantially. Cost per transaction fell.
In the U.S., USDS (United States Digital Service) and 18F have produced pockets of quality government technology — Healthcare.gov after its disastrous launch, the VA’s digital infrastructure, the IRS free file system. These organizations apply the same practices that produce good commercial products: user research, iterative design, performance measurement against user outcomes.
They exist at the margins of a system that, overall, continues to reward the wrong things.
The gap between what is possible and what currently exists in political and government technology is not a technical gap. The knowledge of how to build good digital experiences is widely available and not particularly expensive to apply. It’s an incentive gap — a persistent mismatch between what the systems are optimized to produce and what citizens actually need.
Closing that gap is less a technology challenge than a political one. Which is perhaps why, in an industry that specializes in political problems, it remains largely unsolved.
This is the third installment of "Technology & Democracy," a series about the intersection of technology and democratic systems. Next: "The Ocean Is the Next Geopolitical Battleground."